Assessing the impact of religious organizations on HIV and drug-use prevention requires an understanding of personal religiosity (both inward spirituality and outward manifestations of devotion); of religious organization activtty; and of the network of relationships that connects persons to each other and to organizations. To measure impact, we propose a case-control study (with longitudinal followup) of 80 HIV-positive and 160 HIVnegative persons (divided even by sex) recruited from a stressed inner-city Atlanta neighborhood to ascertain the influence of religious organization activity on HIV status, sexually transmitted diseases (STIs), blood-borne infections (BBIs), and drug use. We will develop a person-time metric for exposure to religious organization HIV preventive activity (e.g., person-hours spent in contact with religious organization representatives). We will assess the impact of such exposure in the context of potential confounding by personal religiosity, network associations, and other demographic and behavioral variables, and will quantify the impact of religious organizations on these outcomes using the Prevented Fraction, an epidemiologic measure of association that specifies the portion of disease prevented that can be attributed to a specific preventive factor or intervention. We will enroll the religious organizations named by respondents, in order to assess organizations' activities, commitment, and connections with each other. A bipartite network (organizations and respondents) will be used to examine the affiliation network of organizations (how they are connected through people). The affiliation network will be compared to the network determined from religious organizations' self reported connections. The gap between these two, which we hypothesize to be considerable, is a measure of the unrecognized potential for intervention that can be achieved by improved interaction of religious organizations. Thus, the logical train that drives the proposal is as follows: Religious organization activity will be significantly associated with HIV-negativity and nondrug-use, but its impact, as measured by the PF, will be small. The interconnections among religious organizations are far greater than they realize.